Transforming LCT Pegmatite Targeting Models into AI-Powered Predictive Maps of Lithium Potential for Western Australia and Ontario: Approach, Results and Implications
Open Access
Abstract
Here, we present holistic targeting models for lithium–cesium–tantalum (LCT) pegmatites in Western Australia, the world’s largest supplier of hardrock lithium ores, and Ontario, an emerging hardrock lithium mining jurisdiction. In this study, the LCT pegmatite targeting models, informed by a review of this deposit type and framed in the context of a mineral system approach, served to identify a set of targeting criteria that are mappable in the publicly available exploration data for Western Australia and Ontario. This approach, which formed the basis for artificial intelligence (AI)-powered mineral potential modeling (MPM), using multiple, complimentary modeling techniques, not only delivered the first published regional-scale views of lithium potential across the Archean to Proterozoic terrains of Western Australia and Ontario, but it also delivered an effective framework for exploration and revealed hidden trends. For example, we identified a statistically verifiable proximity relationship between lithium, gold, and nickel occurrences and confirmed a significant size differential between LCT pegmatites in Western Australia and Ontario, with the former typically containing much larger resources than the latter. Overall, this regional-scale targeting study served to demonstrate the power of precompetitive, high-quality geoscience data, not only for regional-scale targeting but also for the development of camp-scale targets that have the resolution to be investigated using conventional prospecting techniques. Importantly, MPM does not generate ‘treasure maps’. Rather, MPM provides another tool in the ‘exploration toolbox’, and its output should be taken as the starting point for further investigations.Keywords
All Related Versions
This publication has 78 references indexed in Scilit:
- Mineral Prospectivity Mapping based on Isolation Forest and Random Forest: Implication for the Existence of Spatial Signature of Mineralization in OutliersNatural Resources Research, 2021
- FUCOM-MOORA and FUCOM-MOOSRA: new MCDM-based knowledge-driven procedures for mineral potential mapping in greenfieldsSN Applied Sciences, 2021
- Improved index overlay mineral potential modeling in brown- and green-fields exploration using geochemical, geological and remote sensing dataEarth Science Informatics, 2020
- Sustainable supplier selection in healthcare industries using a new MCDM method: Measurement of alternatives and ranking according to COmpromise solution (MARCOS)Computers & Industrial Engineering, 2020
- 3D Mineral Prospectivity Mapping with Random Forests: A Case Study of Tongling, Anhui, ChinaNatural Resources Research, 2019
- An Improved Prediction-Area Plot for Prospectivity Analysis of Mineral DepositsNatural Resources Research, 2018
- Geometric average of spatial evidence data layers: A GIS-based multi-criteria decision-making approach to mineral prospectivity mappingComputers & Geosciences, 2015
- Data-Driven Index Overlay and Boolean Logic Mineral Prospectivity Modeling in Greenfields ExplorationNatural Resources Research, 2015
- Fuzzification of continuous-value spatial evidence for mineral prospectivity mappingComputers & Geosciences, 2015
- Random forest predictive modeling of mineral prospectivity with small number of prospects and data with missing values in Abra (Philippines)Computers & Geosciences, 2015